Core Dimensions

Intro

Reverb is sort of the black arts of signal processing. Most normal signal processing is pure math, which is either right or wrong, and can be proven right or wrong.

Reverbs are different. While there are countless academic papers on modal densities, echo distributions, eigen vectors and other reverb related math and metrics, the bottom line is that the records we listen to use reverbs that are far from perfect. We still haven't seen any "academic" reverbs reach the top of the charts.

The imperfections of a well-designed reverb, and probably more important, the workarounds for those imperfections, are what gives a reverb its character. No-one can argue that a mechanical plate sounds "exactly like a concert hall", or that the modulation of the classic 250 hardware sounds "just like real life". But that's not what it is about, reverb is an artform without rights or wrongs, just like the melody you're about to write.

Core Dimension is a collection of 5 high quality reverb algorithms in one easy to use plug-in. Each algorithm is different and full of character, ranging from analog modeled reverbs to modern digital designs. Designed with speed, flow, and ease of use in mind, Core Dimensions offers a curated set of reverb controls, an intuitive reverb tail visualization, and additional creative effects with Shimmer pitch-shifting and automatic reverb tail ducking.

Understanding Common Reverb Controls

Reverbs are different than most other audio processors, and they have some special terminology. A lot of the terminology was made up when the first reverb processors were built in the seventies and early eighties, and while some terms have disappeared ("mode enhancement", anyone?), others are still in use today.

Here are a few of the most common terms used today.

Time: The length of the reverb tail. Typically measured as the time to when the tail goes below -60 dB (“RT60”), but other reverb controls can change the actual length a lot.

Predelay: The time between the dry signal and the onset of the reverb. Say you want to emulate the sound from a vocalist in a large reverberant space. If the vocalist is close to the mic but far from the walls, it will take a long time for the echo from the walls to arrive at the mic, and that is the time you set with your predelay. In practice, you use predelay to increase the intelligibility of the audio source by separate the dry audio (by 20-40 ms) from the reverb tail.

User Interface

Core Dimensions' main controls are placed in the middle, with Time, Dry/Wet, Dark/Bright and Predelay.

To the right you will find a Ducking control which will automatically "duck" the reverb tail when there is input signal present, a Shimmer effect that can pitch shift the reverb for extra glitter as well as some standard output controls.

To the left is our reverb tail visualizer. We took equal inspiration from the Fairlight CMI waveform display and Joy Division’s Unknown Pleasures cover and created something that intuitively illustrates the sound of the tail. Browse through the presets, listen to the sound, watch the visualization and you'll quickly get the hang of it!

Extended Features Panels

Next to the green Softube logo in the menu bar, you'll find the buttons to open the input, output and lower panels. They are part of Extended Features, a utility which is shared by all Softube plug-ins that contains advanced metering, levels, bypass, and other features. For more information, please see the Extended Features User Manual.

Algorithm Panel (Top)

On the algorithm panel you can choose between five different reverb algorithms, all described in detail in the next chapter. Here's a quick rundown of the different flavors.

Algorithm	Character	Best for	Avoid it on	Bandwidth
Spring	Bright and obvious. It either works or not.	Guitars	Full mix	7500 Hz
Plate Analog	Dark and dense.	Vocals, Drums	Full mix	22 kHz
Plate Digital	Warm and lush.	Vocals, Guitars	Full mix	12 kHz
Synth Hall	Long and lush.	Pads, Synths	Drums	8 kHz
Chamber	Dense and immediate.	Everything, Buses	Nothing	22 kHz

Do not automate the Reverb Algorithm control! It can lead to unwanted noise and CPU spikes.

Center Panel

The center panel contains the four most important controls. These are shared between all algorithms.

Predelay

Sets the amount of delay between the dry audio and the onset of the reverb.

Dark/Bright

Adjusts the overall tonal balance of the reverb via the use of a tilt filter.

Time

Sets the nominal reverb length. All algorithms can do times between 1.1 s and 36 seconds, but some algorithms can go lower or higher. The lowest possible time is indicated by the slider.

Shimmer

Shimmer is an effect where you pitch shift the audio inside of a feedback loop, which means that you can achieve a slowly rising or falling pitch effect. Typically, the classic use case is to pitch shift by a full octave, in order to keep the harmonic integrity of the audio.

The Shimmer effect takes place before the reverbs, which means that it will only affect the incoming audio.

Shimmer

Sets the level of the Shimmer effect. A macro which sets the level of Shimmer effect, and the amount of feedback.

Pitch

Sets the amount of pitch shifting, from -12 semitones (one octave down) to +12 (one octave up).

“Anti-Shimmer” is a beautiful effect where you set Pitch to -12 (down one octave) instead of pitching it up.

Ducking

Ducking is the process of lowering the volume of an audio signal while another (sidechain) signal exceeds a certain threshold and raising it again when the sidechain signal drops back below the threshold.

Threshold

Sets the level the dry/control signal should exceed in order for ducking to occur. You need a relatively low threshold in order for ducking to take effect.

Sidechain

Internal: An internal sidechain keys the dynamics from the dry signal, ducking the wet signal when the dry signal is present.

External: Here you can choose to route an external control signal that will trigger the ducking effect. Please consult your DAW manual for information on how to route an internal signal to the sidechain input in Core Dimensions.

Output Panel

High Cut

6 dB/octave high cut filter, placed after the reverbs.

Low Cut

12 dB/octave low cut filter, placed after the reverbs.

Width

Sets the M/S balance between the mid (mono) and sides. At its minimum setting, the output from the reverb is fully mono, in the center it's fully stereo and at its max setting you only get reverb in the sides, not the mid.

As opposed to most width controls that only allow you to boost the sides by a couple of decibels, the Width in Core Dimensions will completely mute the mid channel when it is at its maximum position.

Width should always be used with care with reverbs. Most reverbs are designed to have a maximum width, and altering the balance between mid and side might lead to side effects, such as resonances or unwanted echoes.

Output Gain

Sets the output level after the Dry/Wet control.

Algorithms

This section provides an overview of the algorithms and their controls.

Spring

The spring reverb in Core Dimensions is tailored from the spring reverb in a classic tape echo unit. It's not as springy as a typical guitar amp spring reverb, and it's not as clean as a high-end studio spring reverb. It's somewhere in between, and it's just a lovely addition to the other algorithms.

Plate Digital

The plate digital algorithm is a recreation of the first commercial digital reverb unit, a big beast with Star Trek-like controls and levers. Despite the original's technical shortcomings, such as very little memory, low sample rate and limited calculations, it is one of the most beautiful reverb designs ever created, and it's been a studio staple ever since.

Plate Analog

While developing the digital plate algorithm we decided to expand on the original concept and bring in some flavor from real mechanical plates. Our model uses resonance modeling to capture the sound of the plates, especially the onset, and carefully tuned internal parameters to recreate the reverb tails of the mechanical plates.

Synth Hall

This algorithm was originally developed as a synth reverb, and different versions of it have been used in our synth plug-ins Parallels and Model 77. In Core Dimensions, we've improved further on the algorithm, especially its modulation, and designed it to be lush and long, without any real sense of room dimensions or reality.

This is the same algorithm as "Cathedral" in the Dimensions plug-in.

Chamber

Chamber is an algorithmic and ideal realization of a Chamber reverb. Through a newly developed reverb tail generation method, and with the power of modern processors, Chambers produces a reverb tail that is maximally dense and modulated without any audible pitch artifacts.

The dense tail of Chamber makes it perfect for short decays (for example drum rooms) where you want to add a depth and thickness to the sound without an audible reverb tail. With long reverb times, the tail of the Chamber algorithm sounds very different from the tail of Synth Hall. Chamber will preserve the pitch of the audio while Synth Hall (and other modulated reverbs) will diffuse the pitch into a noise-like mush after a couple of seconds. Both variants are beautiful sounding, but different.

Block Diagram

This block diagram outlines the basic structure of the Core Dimensions plug-in.

Credits

Niklas Odelholm – Product design, reverb modeling, presets, manual Cameron Clark – Shimmer and Chamber programming, UI programming, presets, manual Johan Bremin – Presets, quality assurance Anna Pohl Lundberg – Reverb visualization Kim Larsson – DSP programming Anton Eriksson – DSP programming